Pet tracker Z8 - help please identifying protocol

Neil4 years ago

your cell data returns this in google's GeoLocation API.

{
  "location": {
    "lat": 48.886438299999995,
    "lng": 2.0917778
  },
  "accuracy": 863
}

accuracy is in metres.

The CID,LAC are in hex, not sure if the Decibel level is in hex or decimal, i tried both, got the same location result from google.

Phil4 years ago

Hi,
I confirm your analysis. I could check that the MCC/MNC/LAC/CID information returned by the tracker corresponds to BS in my neighborhood (surprinsingly this small thing detects a lot of BS).
The value following the 'E' in the GPS measurement is the speed (0.00 in the example below. I was static), probably expressed in km/h
The value which follows (19.79 in the example below) is very likely related to accuracy (URE, GDOP), which depends on the time of the day and the location. Difficult to check...
I have installed netmonitor on my phone to check the receive power. It has to be compared at the same location and same time, but it is normal to observe some differences, as the receivers are different in the tracker and the phone (antenna gains....)
Here is an example of a message returned by the tracker. It is the only one containing both LBS and GPS data!

Tue Nov 3 09:24:07 2020 Message from tracker:

b'PC,3512IMEI#######,1{<U103#082406,A,4853.2988,N,00204.9771,E,0.00,19.79,031120,3;;7,EB2,97,35,EB1,97,30,F11,97,20,EB0,97,17,7F20,97,12,F12,97,9,4E49,97,4,14,D0;031120,082406|95|FFFFFFFF>}\r\n'

Here is an indented version:

b'PC,3512IMEI####,1
    {<U103#082406,A,4853.2988,N,00204.9771,E,0.00,19.79,031120,3; #time, 'A', GPS N, GPS E, Speed, GDOP/URE ?, day, Number of satellites ?
        ;
        7,				# number of BS measured
            EB2,97,35,		# BS1: Cell Id, LAC, RSSI (dBm?)
            EB1,97,30,
            F11,97,20,
            EB0,97,17,
            7F20,97,12,
            F12,97,9,
            4E49,97,4,
        14,D0;				  # MCC, MNC (France, Bouygues telecom)
        031120,082406|95|FFFFFFFF>}\r\n'  # day, time , battery level, bitmap

We are not very far from being able to impersonate the china server ...

Neil4 years ago

"returned by the tracker corresponds to BS in my neighborhood (surprinsingly this small thing detects a lot of BS)"

that made me laugh.. so you have plenty of BS around your area.

joking aside.

"the value following the 'E' in the GPS measurement is the speed (0.00 in the example below. I was static)"

The value following E..... with 0.00, i'd expect it to be an estimated speed... if you are allowed out, take it for a bike ride and log the results or hold it above your head in your hand making a big figure 8 multiple times.. that sometimes wakes them up if idle...check speed then... compare to app screen.

19.79 - i expect to be a estimated accuracy in metres of the current A/V GPS location, check the app screen for a circle around position marker or even if it's shown as a value.

A field is an actual GPS location, you'll see a V if estimated position that follows... that's what i've observed on many of the trackers i have.. the date and time of the actual GPS is obvious.

3 i would expect to report the number of satellites, you normally need a min of 4 for an actual proper fix, 3 for trilateration (2D the point of intersection of 3 circles) and 1 for altitude... to provide the 3D position (x,y + z).

The only block you are missing which the Z8 would always return ;; between the GPS and LBS is the wifi data...same protocol for the F9 & G9 trackers (both wifi).

the 4 byte bitmap i've observed on other trackers has bit settings for things like fuel low, ignition disabled.. etc.... check this.. put it on charge and see if the bits changes.. that's another one i observed.. when on charge and not... one of the bits will change if mapped.

Neil4 years ago

following on from my last reply:

If you sign up to Geolocation API.

This will give you the API usage for querying Wifi and/or LBS tracker data:
https://developers.google.com/maps/documentation/geolocation/overview

A call to: https://www.googleapis.com/geolocation/v1/geolocate?key=API_KEY

passing json:

{
  "homeMobileCountryCode": 208,
  "homeMobileNetworkCode": 20,
  "radioType": "gsm",
  "considerIp": "false",

  "cellTowers": [
    {
      "cellId": 3761,
      "locationAreaCode": 151,
      "signalStrength": 57,
      "mobileCountryCode": 208,
      "mobileNetworkCode": 20
    },
    {
      "cellId": 3760 ,
      "locationAreaCode": 151,
      "signalStrength": 66,
      "mobileCountryCode": 208,
      "mobileNetworkCode": 20
    },
    {
      "cellId": 3762,
      "locationAreaCode": 151,
      "signalStrength": 40,
      "mobileCountryCode": 208,
      "mobileNetworkCode": 20
    },
    {
      "cellId": 3857,
      "locationAreaCode": 151,
      "signalStrength": 33,
      "mobileCountryCode": 208,
      "mobileNetworkCode": 20
    },
    {
      "cellId": 20042,
      "locationAreaCode": 151,
      "signalStrength": 19,
      "mobileCountryCode": 208,
      "mobileNetworkCode": 20
    },
    {
      "cellId": 3858,
      "locationAreaCode": 151,
      "signalStrength": 3,
      "mobileCountryCode": 208,
      "mobileNetworkCode": 20
    }
  ]
}

returned:

{
  "location": {
    "lat": 48.886640299999996,
    "lng": 2.0917301999999998
  },
  "accuracy": 868
}
Phil4 years ago

Thanks for your feedback. This Google API is a good hint. I need to get more familiar with it
The GSM spectrum will progressively refarmed to 4G (and 5G), but the moment the GSM coverage is still there.

Neil4 years ago

Unless they've give you a timescale, it won't be at least for a few years yet (they haven't started doing it here in the UK).

I've just go my hands on a new 4G (with 3G & 2G) tracker, i'm going to give that a test run within the next few days.. i do have the protocol documentation for that.

Phil4 years ago

Hi Neil,
I would like to share my latest findings (on speed, accuracy and number of satellite). I have a .html file opening an OpenStreetmap map (we can see the speed, number of satellites and accuracy on each point)
I was driving my scooter at a speed of about 22km/hour.
It looks like the speed is expressed in miles/hour
The unit used for accuracy is strange. there is a strong correlation between the number of satellites (8-12) and the accuracy. But the values are not expressed in meters, neither miles (?)
I would like you to have a look to it. Do you know if the forum allows the transmission of files in a one to one mode ?

I have also ordered another tracker supporting 4G and WiFI (this one being no longer usable. the Android application is faulty, not the tracker which reports measurements every 10 second)

Neil4 years ago

If you want to take this off-line and contact me directly, go to Critter Alert dot com and contact me via the contact page.

Neil4 years ago

is it in Chinese miles?
1 li to miles = 0.31069 miles.

they're normally in km... unless you haven't set the language and timezone (sms command above for lang) check app settings.

the app gets stuck in refresh mode and becoming borderline useless.

my 4g tracker arrived last week, just been testing it.

i have about 40 trackers, pretty much all the latest ones, i've done a lot of testing... as this related to the Z8, instead of polluting this topic further, you can contact (prior message)

Neil4 years ago

@Phil, I never heard back from you, any luck with the distances (mi/km/li)?

Phil4 years ago

Hi,
You discouraged me to continue writing on the forum :-)
I found the response: the speed is expressed in nautics miles (185.2m/h) more exactly in 'encablures' (I did not find the english translation). Encablure is one tenth of a mile. I compared with the speed measured on my scooter. It matches perfectly.

I am now playing with Android to develop the client application in replacement of the vendor's application (just for the fun)

Also, big disappointment with my new tracker RF-V43 (WiFI+4G), it never locks on satellites. The Alibaba vendor name is Prazata. they sell only trackers, but are totally incompetent in their understanding of how mobile networks as well as GPS work (I have on my side >20years of experience on mobile communications in 5G, also involved in high accuracy positioning for 5G)

Neil4 years ago

Phil: you never responded back via email.. i need to work with you on the RF-V43, i've got 4 and issues as well... that's not related to this topic.. contact via email we'll discuss it there.

Phil4 years ago

Hi Neil,
I contacted you twice (the first one was several weeks ago) on your web site. You never responded.
I am interested to discuss and work on the RF-V43. I received the second one now, it seems to work much better than the first one. I need to make more serious tests in the forest with my dog (to see the sensitivity, and how it reacts when it is back from an uncovered area)
Best,
Philippe

Neil4 years ago

Philippe, i thought you were ignoring me.
I responded to your outlook account and your a few to your Gmail at the start.

OverkillTASF4 years ago

I'd appreciate if y'all could keep your discussion, or at least very specific updates on your findings and progress, here. I'm finding it informative as an observer. Also... I'm looking for a device like this and y'all's reviews are valuable.